AI Prompt Engineer

Description

The company is hiring an AI Prompt Engineer for a full-time role on a team building AI solutions for U.S. Customs and Border Protection (CBP).

The work uses generative AI and large language models to help frontline personnel find information, make decisions and complete tasks more efficiently, with a focus on situational awareness and fast access to mission-critical information.

The role sits within the company's Agile Solution Factory, a group that runs rapid experiments and prototypes and values continuous learning.

The team wants people who are curious about how AI capabilities are changing, and who can turn complex operational problems into practical AI solutions that deliver measurable value to mission users.

Responsibilities

• Write, refine and iterate on prompts for different generative AI models to get the behaviour, answers and creative output you need.
• Test prompt variations side by side, analyse the results and use them to keep improving.
• Work with several groups to build prompts into workflows that cut down manual processes.
• Keep up with new developments in LLMs, generative AI and prompt engineering methods.
• Contribute ideas and help plan approaches to model fine-tuning and Retrieval Augmented Generation (RAG) that raise accuracy.
• Diagnose unexpected model behaviour and find the root cause, whether it lies in the prompt structure, the model's limits or the data context.

Requirements

• You must be a U.S. citizen who can pass a CBP background investigation. The checks include, but are not limited to, a 3-year review for felony convictions, a 1-year review for illegal drug use, and a 1-year review for misconduct such as theft or fraud.
• You must be able to work onsite in Ashburn, VA and/or Washington, D.C.
• You need a bachelor's degree in computer science, software engineering, information technology or a related field. Equivalent experience may be considered in place of the degree.
• You need 1+ years of software development experience, through work, internships or relevant academic projects.
• You need 1+ year of hands-on work with large language models (for example OpenAI GPT series, Llama or Gemini) or other generative AI models.
• You need experience in at least one programming language, such as Python or JavaScript/TypeScript.
• You need familiarity with prompt engineering techniques such as few-shot prompting, chain-of-thought, structured outputs and function calling.
• Desired:
• Problem-solving skills and a data-driven approach to testing and optimisation.
• Clear written and verbal communication, including explaining complex technical ideas simply.
• Ability to work on your own and with others in a fast-paced, changing environment.
• Exposure to RAG concepts, including embeddings and vector databases.
• Understanding of AI safety issues such as prompt injection, bias and data privacy.
• Exposure to Agile or Scrum development environments.

Conditions

Full-time position.

Onsite work in Ashburn, VA and/or Washington, D.C.

Travel of up to 10%, local only.

No security clearance is required to start.

The proposed salary range is $61,100 - $122,200.

Final pay depends on factors including location, Federal Government contract labor categories and wage rates, relevant prior experience, specific skills, education and certifications.

The company provides equal opportunity in employment.

Benefits

• Comprehensive benefits, including healthcare, wellness, financial, retirement, family support, continuing education and time off.
• A flexible time off benefit that lets you take the time you need.
• Access to learning resources for professional development.